INDEX

Explanations

named followed by backtick

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 магазина

2.05

 некоторых

2.04

 تمام

1.99

 منع

1.95

 обсу

1.93

 другого

1.93

 защиту

1.91

 άλλο

1.90

 سرعت

1.90

 другую

1.89

POSITIVE LOGITS

shorts

1.49

 communément

1.46

 auxili

1.43

 patronage

1.42

oretically

1.41

 simply

1.39

 mere

1.37

之为

1.36

ucos

1.34

 reff

1.33

Activations Density 0.119%